tokenizer_config.json

{
  "additional_special_tokens": [
    "[NEWLINE]",
    "[URL]",
    "[ID]",
    "[EMAIL]",
    "[SN]",
    "[EN]"
  ],
  "clean_up_tokenization_spaces": true,
  "cls_token": "[CLS]",
  "do_lower_case": false,
  "mask_token": "[MASK]",
  "model_max_length": 512,
  "pad_token": "[PAD]",
  "sep_token": "[SEP]",
  "strip_accents": null,
  "tokenize_chinese_chars": true,
  "tokenizer_class": "DistilBertTokenizer",
  "unk_token": "[UNK]"
}